 Open Access
 Total Downloads : 366
 Authors : R. K. Verma
 Paper ID : IJERTV3IS061548
 Volume & Issue : Volume 03, Issue 06 (June 2014)
 Published (First Online): 01072014
 ISSN (Online) : 22780181
 Publisher Name : IJERT
 License: This work is licensed under a Creative Commons Attribution 4.0 International License
Theoretical Modelling of An End Face Reflection based Tapered Fiber Optic SPR Sensor with Different Probe Designs
Rajneesh. K. Verma Department of Physics Central University of Rajastha Bander Sindri, Ajmer, India
Abstract – The theoretical modelling of a SPR based tapered fiber optic sensor is presented with the end face as a reflecting surface. The performance of the sensor is evaluated in terms of its sensitivity and detection accuracy with taper ratio. Apart from this, the effect of various taper profiles on to the sensitivity of the sensor is also studied. It has been found in the present study that the end face reflection technique reduces the size of the sensor probe and makes it miniaturized. The maximum sensitivity of the exponential taper profile makes it an optimum taper profile among the three profiles named linear, parabolic and exponential.
Keywords – Surface Plasmons, Fiber Optics, Tapered Fiber, Plasmons

INTRODUCTION
Surface plasmons resonance (SPR) is a very promising tool
exponential linear has also been studied and a comparative analysis has also been provided in terms of the best taper profile for the sensitivity among the rest of the two.

THEORETICAL MODEL
The theoretical model is based on the principle of attenuated total reflection (ATR) with Kretschmann configuration. The plastic cladding from the middle portion of a step index multimode fiber is removed and the unclad portion is tapered. The unclad tapered region is then cut from the point where the waist diameter is minimum. The unclad tapered fiber region is then coated with a thin gold layer, which is further surrounded by sensing medium (Fig. 1).
Sensing Region
for the sensing of various biochemical and gaseus sensing [1 12]. A vast number of researchers are engaged in this upcoming area to explore various possibilities in terms of making fiber optic probes for the continuous monitoring of changes occurring in biochemical, chemical and gasesus media [1227]. Recently many theoretical models have been
2i
Cladding
Core
20
Reflecting Surface
proposed for the better sensing mechanisms such as localized surface plasmon based fiber optic sensors [3], use of conducting metal oxides is also an important study [10][14]. In the last year people have fabricated and characterized the fiber optic sensing probes for pH sensing [10] and H2S gas sensing [11]. Sensing of Phenolic compounds and Ammonia gas sensing has also been proposed [12][13]. Fiber optic grating is also being used in conjunction with spr to make a miniaturized fiber optic probe for sensing the refractive indecies [14][15]. Most recently people are also proposing certain models on optical waveguides which involves artificially structured materials such metamaterials as a SPR active material [16]. In this study a theoretical model has been proposed which involes the multimoded tapered optical fibers. Contrary to the conventional fiber optic probes, in the present probe the one of the end face is coated with a silver paste to make it reflecting. This probe proposed in the model will be miniaturized as well as it will work as a point sensor as its one end can be dipped directly in to the medium to which we are going to sense. The performance parameters such as sensitivity have also been evaluated in terms of the taper ratio. Effect of various taper profiles such as linear, parabolic and
Gold layer
L
Figure 1: A typical SPR based fiber optic sensor with tapered probe.
Different taper profiles can be generated during fabrication using Flame Brush technique as suggested by Birks and Li [23][24]. The method uses a point like heat source (a gas burner) which heats only a small section of the fiber at a time. This burner is made to travel at constant speed in an oscillatory manner along the distance of the fiber (to be tapered) so that in each cycle of oscillation every element of the fiber is heated identically. If the speed of the burner is large compared to the tapered elongation speed then a time averaged hot zone is set up in the fiber which satisfies the condition that the fiber section being heated is always cylindrical and is always heated uniformly. Allowing the hot zone length to change as taper elongation precedes any shape of the fiber taper can be generated. Three taper profiles namely linear (LP), parabolic (PP), and exponential linear
(EP) have been studied in literature for a fiber optic evanescent field absorption sensor as shown in Fig. 2 [22]. We consider these profiles for SPR based fiber optic tapered sensor. The different parts of the sensor are discussed below.
Figure 2: Three different taper profiles showing the linear, parabolic, and exponentiallinear variation of core radius of taper with distance z from the
C. Reflection Coefficient (Rp):
We consider Nlayer model to obtain the expression for amplitude reflection coefficient for ppolarized incident beam as shown in Fig. 3 [5].
Figure 3: NLayer model to determine the reflected light intensity.
In the present study, N is equal to 3. The layers are assumed to be stacked along the zaxis. The arbitrary medium layer is defined by thickness dk, dielectric constant k, permeability k, and refractive index nk. The tangential fields at the first boundary Z = Z1 = 0 are related to those at the final boundary Z = ZN1 by
input end of the taper.
U U

Core of the optical fiber
1 M
N 1
For theoretical modeling, the optical fiber considered is a stepindex multimode plastic clad silica fiber. Sinced the
V1
VN 1
(3)
refractive index of the silica varies with the wavelength therefore the wavelength dependence of the refractive index (n1) of the silica core is given by Sellmeier dispersion relation:
In Eq. (3) U1 and V1 are the tangential components of electric and magnetic fields at the boundary of first layer, respectively; UN1 and VN1 are the corresponding fields at the boundary of Nth layer and M, known as characteristic matrix of the combined structure, is given by
M
n1
2
a
1 1
2 b2
2
a
2
2 b2
2
a
3
2 b2
N 1
M
M11
M12
1 2 3
(1)
M k
k 2
21 M
22
(4)
a1, a2, a3, b1, b2, and b3 are the Sellmeier coefficients and
is the wavelength of light used in m. The values of these coefficients are given in the table below [25]:
with
M
cos k
i sin k
/ qk
k
a1
0.6961663
a2
0.4079426
a3
0.8974794
b1
0.0684043
b2
0.1162414
b3
9.896161
where
k iq
sin k
cos k
(5)
1/ 2
n2 sin 2
1/ 2
q k cos
k 1 1

Dispersion relation for plasmonic material
For the wavelength dependence of the dielectric
k k
k
k
(6)
constant of the plasmonic material we have used the Drude model, given as
2 n
k k
cosk
zk

zk 1
2dk
k
n2 sin 2
1/ 2
1
1
m mr

imi
1
2
c
2 i
(7)
p c (2)
where p and c denote the plasma wavelength and collision wavelength, respectively. For gold, the following values of the plasma wavelength and collision wavelength are used: P= 1.6826Ã—107 M AND C= 8.9342Ã—106 M [26]
and k is the angle of the ray normal to the kth interface. The amplitude reflection coefficient (rp) for ppolarized incident wave is given by
p
r M11 M12qN q1 M 21 M 22qN
Exponential linear:
z
e L e1
0
M11

M12qN
q1
M 21

M 22qN
(8)
L
z z
e i 0
(13)
Finally, reflectance (Rp) for ppolarized light is
2
Rp rp
Parabolic:
2 z 2
2 1/ 2
(9)
.
P (z) i

L i
0
(14)

Reflected Power
The light is launched into the one end of the fiber through a broadband light source with proper optics. The other end is polished with silver paste to make it reflecting. The light rays that travels to the fiber gets reflected back due to this mirror end face. The end through which light is launched is connected with two way reflection probe. One end is used for
If (1, 2) is the range of incident angles of the rays launched at the input end of the taper, then at a distance z, this range alters to [1(z), 2(z)] due to variation in core diameter in tapered region and the transmitted power at the output end of the fiber is given by
light launching and other is used for collecting light back. This reflecting end is connected to a detection system to record the
L 2 ( z )
dz
2 N ( , z ) n 2 sin cos
ref 1
R d
output signal. Consider the propagation of all the guided rays
P
(1 n 2 cos 2 )2
launched in the fiber using a collimated source and a
P 0 1 ( z ) 1
microscope objective. The objective focuses the beam at the
trans
L 2 ( z )
n 2 sin cos
endface of the fiber at axial point. The angular power distribution of rays guided in the fiber with as the angle of the ray with the normal to the corecladding interface is given by [22]
dz
0
1 ( z )
1
2
(1 n1
d
cos 2 )2
(15)
n2 sin cos
dP 1 d
The expressions for 1(z) and 2(z) can be found by
substituting the value of in Eq. (11). For 1(z),
1
1 n2 cos 2 2
sin
ncl
n1 and for 2(z),
2 . Further, Nref
1 (10)
To determine the effective reflected power, the reflectance (Rp) for a single reflection is raised to the power of the twice the number of reflections the specific propagating angle
is the number of reflections the ray of angle undergoes in the fiber with radius (z) in the length L and is given as
undergoes with the sensor interface (Due to reflecting end each ray retraces its path). The guided rays in the uniform core fiber enter the sensing tapered region as shown in Fig. 1. Assuming adiabatic (slow) tapering, if (z) is the taper radius
Nref
( , z)
L
2(z) tan
(16)
at a distance z from the input end of the taper then the angle of the ray will transform into angle (z) in the tapered region and is given by [22]

Sensitivity
In any fiber optic SPR sensor based on wavelength interrogation, the light from a polychromatic source is launched into one of the ends of the fiber and the spectrum of the transmitted light received at the other end is recorded to
1 i cos
1 i 0
know the resonance wavelength (the wavelength
z cos
z
tan
L
(11)
corresponding to minimum transmission) corresponding to the refractive index of the sensing layer. If sensing region has a
In above equation, i and 0 are, respectively, the radii of the input (i.e. z = 0) and output end (i.e. z = L) of the taper. Further, the second term on right hand side is the taper angle () and L is the taper length. The taper radius varies with z in different fashion for different profiles given as [17]
refractive index value of ns and m is the dielectric constant of the metal layer in contact with sensing region, then the surface plasmon resonance occurs when the following condition is satisfied
Linear:
2 2
n 2
1 / 2
z n sin Re
m s
(z) i i 0
1
n 2
L (12)
m
s
(17)
In Eq. (17), the expression on the left hand side is the propagation constant (Kinc) of the light incident at an angle while the right hand side represents the real part of surface
plasmon propagation constant (KSP). In the SPR sensor based b. on wavelength interrogation, Fixed angles of incidence are
used and resonance wavelength (res) is determined corresponding to refractive index of the sensing layer (ns). If the refractive index of the sensing layer is altered by ns, the resonance wavelength shifts by res. The sensitivity (Sn) of a SPR sensor with spectral interrogation is thus defined as [5]
Sn
res
n
s


RESULTS AND DISCUSSIONS
(18)
Various important parameters used in the present study have been taken as: numerical Aperture of the fiber = 0.22, diameter of the fiber core (2i) = 600 m, sensing region length (L) = 0.5 cm, gold layer thickness = 50 nm. In order to analyze the effect of taper ratio and taper profiles on the sensitivity, first of all we have to analyze the plasmonic resonance condition given by Eq. (17). As it is evident from Eq. (17), propagation constant (KSP) of surface plasmon wave (SPW) depends on the wavelength of the light as well as on the dielectric constant of metal layer (m) and refractive index (ns) of sensing region. Whereas, the propagation constant (Kinc) of the incident wave depends on the angle and wavelength of the light ray as well as on the refractive index of the fiber core. Figure 4a depicts the variation of KSP1 (i.e. for ns1 = 1.333), KSP2 (for ns2 = 1.343) and Kinc with wavelength for the case when no tapering of the fiber probe is done. The angle of the ray is taken at the middle of the corresponding angle (1, 2). As can be seen, KSP and Kinc curves cross each other at 613.86 nm for ns1 = 1.333 and at
648.46 nm for ns2 = 1.343. Thus, a shift (res) of 34.60 nm in resonance wavelength is observed for ns = 0.010. Now, the next task is to estimate the effect of taper profile (Eqs. 12
14) and taper ratio (i/o ) on the shift (res) for the same value of ns. Figure 4b shows the variation of KSP1, KSP2, Kinc (LP), Kinc (PP) and Kinc (EP) with wavelength for taper ratio of 3.0. The curves corresponding to KSP1 and KSP2 are similar to those in Fig. 4a.
a.
Figure 4: Variation of propagation constants for SPW and incident wave with wavelength for (a) Taper ratio = 1.0 and (b) Taper ratio = 3.0.
A noticeable change in Fig. 4b can be seen in terms of three fold variation of Kinc corresponding to three different taper profiles. This happens due to two reasons: one, the angular range (1, 2) is different for different taper profiles, and second, the dependence of both 1 and 2 on the taper ratio (Eq. 11). Consequently, the curves corresponding to KSP1 and KSP2 will intersect these three Kinc curves at six different points. In Fig. 4b, the above intersections corresponding to LP, PP, and EP have been shown with the help of two each solid, dashed, and dotted arrows, respectively. Table below shows the points ofintersection and corresponding shift (res) for three taper profiles and three taper ratios.
It is quite visible from Fig. 4b and from Table above that for a given ns, maximum shift (i.e. res) is obtained for EP whereas minimum shift is found for PP. For LP, the corresponding shift is found to be in between the two. The similar trend was observed for all profiles at several angles in the corresponding range (1, 2) and at several taper ratios between 1.0 and 2.2. This observation prepared the ground to check the present case when all the guided rays are launched together in the tapered fiber.
The calculations for transmitted power for a given refractive index of the sensing region, using Eq. (15), were
carried out with all the guided rays launched separately for all the three taper profiles with different taper ratios. From the plot of SPR spectrum resonance wavelength (res) was determined. To appreciate the shift in resonance wavelength (i.e. res) due to the change in the refractive index (ns) much more precisely, the refractive index change (ns) was narrowed by taking the values of ns1 and ns2 as 1.333 and 1.335, respectively, (i.e., ns = 0.002). As was expected from above discussion, the shift (res) was found to be maximum for EP and minimum for PP with LP lying in between the two. The sensitivity of the sensor was calculated as defined by Eq. (18). Figure 5 gives the corresponding variation of the sensitivity (in m per RIU) with taper ratio for all the three taper profiles. The figure shows that the sensitivity increases with an increase in taper ratio for any taper profile. Further, the sensitivity is maximum for exponentiallinear profile (EP). The minimum sensitivity is observed for parabolic profile (PP). For the present set of design parameters mentioned above, a significant sensitivity enhancement of more than 80% in comparison to untapered fiber optic SPR sensor is observed at a taper ratio of 3.0 for exponentiallinear taper profile (EP).
Figure 5: Variation of sensitivity (in m per RIU) with taper ratio for three different taper profiles.
Apart from this explanation of the sensitivity enhancement with taper profile and taper ratio in terms of the shifting of resonance point (i.e., KSP = Kinc), another physically logical reason can be given in terms of evanescent wave absorption. By tapering the fiber we intend to broaden the angular range from (1, 2) to (1, 2), because lower value of angle means more spreading of the evanescent field into the region adjacent to fiber core. Hence the coupling between evanescent wave and SPW will become stronger and sensitivity is bound to increase due to tapering. Now, as far as the effect of taper profile is concerned, the angular range (1,
2) in tapered region will be maximum for EP and minimum for PP according to Eqs. (13) and (14). This is why sensitivity is found to be maximum for EP. Furthermore, this improved coupling between evanescent wave and SPW due to tapered fiber causes SPR curve to broaden. However, the degree of SPR curve broadening is much lesser than the sensitivity enhancement with taper ratio and taper profile.

CONCLUSIONS
In summary, a comprehensive theoretical model of an end face reflecting tapered fiber optic probe has been presented to evaluate the influence of tapering on the performance of a fiber optic SPR sensor. It is found from the study that the sensitivity of the sensor increases quite significantly with taper ratio. Also, the use of exponentiallinear taper profile instead of linear or parabolic taper profile further improves the sensitivity. Therefore, high taper ratio with exponential taper profile is recommended to design a highly sensitive fiber optic SPR probe. The light launching optics is also used for collecting the reflected light therefore it is more miniaturized as compared to other fiber optic probes.
REFERENCES
[1]. R. C. Jorgenson, S. S. Yee, A fiberoptic chemical sensor based on surface plasmon resonance, Sens. Actuators B vol. 12 pp. 213 1993. [2]. O. Esteban, R. Alonso, C. Navarrete, C. A. G. Cano, Surface plasmon excitation in fiberoptic sensors: a novel theoretical approach J. Lightwave Technol. Vol. 20 pp.448 2002. [3]. L. K. Chau, Y. F. Lin, S. F. Cheng, T. J. Lin, Fiberoptic chemical and biochemical probes based on localized surface plasmon resonance, Sens. Actuators B vol.113 pp.100 2006. [4]. S. A. Zynio, A. V. Samoylov, E. R. Surovtseva, V. M. Mirsky, Y. M. Shirsov, Bimetallic layers increase sensitivity of affinity sensors based on surface plasmon resonance, Sensors vol. 2 pp.62 2002. [5]. A. K. Sharma, B. D. Gupta, On the performance of different bimetallic combinations in surface plasmon resonance based fiber optic sensors,J. App. Phys. vol. 101 093111 2007. [6]. A.K. Sharma, B.D. Gupta, Fiber optic sensor based on surface plasmon resonance with AgAu alloy nanoparticle films, Nanotechnology vol. 17 pp.124 2006. [7]. G. G. Nenninger, P. Tobiska, J. Homola, S. S. Yee, Long range surface plasmons for high resolution surface plasmon resonance sensors, Sens. Actuators B 74 pp. 145 2001. [8]. Rajan, A. K. Sharma, B. D. Gupta, Fiber optic sensor based on long range surface plasmon resonance, J. Opt.: Pure and App. Opt. vol.9 pp. 682 2007. [9]. A. K. Sharma, Rajan, B. D. Gupta, Influence of dopants on the performance of a fiber optic surface plasmon resonance sensor, Opt. Comm. vol.274 pp. 320 2007. [10]. S. K. Misra and B. D. Gupta, Surface plasmon resonance based fiber optic pH sensor utilizing Ag/ITO/Al/hydrogel layers, Analyst vol.138 pp. 2640 2013. [11]. R. Tabassum, S. K. Misra and B. D. Gupta, Surface plasmon resonancebased fiber optic hydrogen sulphide gas sensor utilizing Cu ZnO thin films Physical Chemistry Chem. Phys. vol.15 11868 2013. [12]. S. Singh, S. K. Misra and B. D. Gupta, SPR based fibre optic biosensor for phenolic compounds using immobilization of tyrosinase in polyacrylamide gel, Sensors and Actuator B vol. 186 pp. 388 2013. [13]. S. K. Misra, D. Kumari and B. D. Gupta, Surface plasmon resonance based fiber optic ammonia gas sensor using ITO and polyaniline, Sensors and Actuator B vol. 171172 pp. 976 2012. [14]. Y.J. He, Y.L. Lo, J.F. Huang, Optical fiber surface plasmon resonance sensor employing long period grating multiplexing, J. Opt. Soc. Am. B vol. 23 pp. 801 2006. [15]. G. Nemova, R. Kashyap, Fiber Bragg grating assisted surface plasmon polariton sensor, Opt. Lett. vol.31 2118 2006. [16]. X. Chen Â· R.R.Wei Â· M. Shen Â· Z.F. Zhang Â· C.F. Li, Bistable and negative lateral shifts of the reflected light beam from Kretschmann configuration with nonlinear left handed metamaterials, Applied Physics B vol.101 pp. 283 2010. [17]. J. Villatoro, D. MonzonHernandez, E. Mejia, Fabrication and modeling of uniform waist single mode tapered optical fiber sensor, App. Opt. vol. 42 pp. 2278 2003. [18]. A. Diez, M. V. Andres, J. L. Cruz, In line fiber optic sensors based on excitation of surface plasmon modes in metalcoated tapered fibers, Sens. Actuators B vol. 73 pp. 95 2001. [19]. Y. C. Kim, W. Peng, S. Banerji, K. S. Booksh,Fiberoptic surface plasmon resonance for vapor phase analyses, Opt. Letters vol 30 pp.2218 2005. [20]. F.J. Bueno, O. Esteban, N.D. Herrera, M.C. Navarrete, A.G. Cano, Sensing properties of asymmetric double layer covered tapered fibers, App. Opt. vol. 43 pp. 1615 2004. [21]. B. Grunwald, G. Holst, Fiber optic refractive index microsensor based on white light SPR excitation, Sens. Actuators A vol. 113 pp. 174 2004. [22]. B. D. Gupta, C. D. Singh, Fiber optic evanescent field absorption sensor: effect of launching condition and the geometry of the sensing region, Fiber and Integrated Optics vol.13 pp. 433 1994. [23]. G. Chen, Profile optimization of tapered waveguide sensors by fluorescence imaging, Proc. SPIE 5589 (2004) 70. [24]. T.A. Birks, Y.W. Li, The shape of fiber tapes, J. Lightwave Technol. Vol.10 pp. 432 1992. [25]. A.K. Ghatak, K. Thyagarajan, Introduction to Fiber Optics, Cambridge University Press, 1999, 85. [26]. M.A. Ordal, L.L. Long, R.J. Bell, S. E. Bell, R.R. Bell, R.W.
Alexander Jr., C.A. Ward, Optical properties of the metals Al, Co, Cu, Au, Fe, Pb, Ni, Pd, Pt, Ag, Ti, and W in the infrared and far infrared, Appl. Opt. vol 11 pp.1099 1983.